Skip to content
Error

Failed to copy link to the clipboard.

Success

Link copied to the clipboard.

Writing Talking Interfaces

Talking interfaces in academia

Up until this point, Talking Interfaces has presented itself formally as a conventional academic essay (albeit published on a webpage). What follows, however, is an experiment in using talking interfaces within the specific arena of academia.

In the previous section this project conformed to many of the expectations of an academic article or book chapter. It introduced the subject and some definitions, acknowledged the scholarship it builds upon, and made some claims as to how it might expand knowledge and contribute to current debates. The language and syntax were relatively formal, and the prose attempted to outline a linear argument via statements and the occasional question. The form was also familiar, subtly adapted for the digital platform (with hypertextual links and embedded content) but not outlandishly so. The footnotes, if clunky, are still there. So too are section headings.

In the remainder of this contribution I seek to put pressure on some of these conventions. The structure itself explores how academic writing might be developed within a digital context and an era dominated by talking interfaces – a meta-methodological contribution if you will.

Though one with caveats.

I do not present a conversational agent. In a perfect world the reader would have talked with a conversational agent here, utilizing a talking interface as a means of maneuvering through this scholarship. The process of interacting with that conversational agent would, I hoped, illustrate the affordances and limitations of talk and one particular talking interface for the reader in an engaging and thought-provoking way. Interactive nonfiction or interactive orature so to speak. However, such an interface would have required financial resources, platform affordances, and expertise above those I possess and would have presented significant preservation challenges (how do you archive a talking interface?) that are outside of the scope of the Ego-Media project. I hope to consider some of these challenges in future research.

A second alternative might have been to present a chatbot. Utilizing writing (typing) rather than speech in its interface, a chatbot nevertheless purports to offer the human user a conversation, or chat, as a means of interacting with the software. Based on natural language processing, chatbots have been popular since the 1960s, offering talk interfaces in the years before speech recognition software was widely available or effective. For scholars of literature and life writing, schooled in the written word, chatbots are appealing thanks to their mediation of speech into writing. More widely, they purport to mediate talk via writing via a talking interface. In so doing, they can help to reveal the assumptions we make concerning the value or otherwise of these different technologies. The recent explosion of social media applications that purport to enable “chat” via text indicates the degree to which this mediation is not limited to the chatbot. The relative longevity, exemplarity, and mediative complexity of chatbots has led me to focus my analysis on this particular kind of talking interface in what follows. (For correlative work on voice, see Rob Gallagher’s work)

Unfortunately, building a chatbot as a means of interacting with this content creates many of the same issues identified above (although I do discuss building one chatbot, EgoBot). Instead, I adopt the underlying structure which governs chatbots – and indeed all programming – namely logic gates. A simple chatbot models a conversation into a series of defined options for progressing to the next interaction.1 In so doing it conceives of talk in a very specific manner: as in essence a sequence of choices in which the user’s agency is constrained by the knowledge base and programming of the machine interlocutor. Here I present the reader with a similar series of choices over which section of material they wish to examine next. Talking Interfaces is thus navigable via multiple routes—a kind of academic “choose your own adventure” in which the reader’s options are prelimited while also engendering a kind of interactivity.2,3,4 It is a modest example of what might be termed hypertextual, ludic, or what performance studies scholar and practitioner Peggy Phelan has called “performative” writing.5,6,7,8,9

Talking Interfaces also exposes the conventions of the humanities-based academic monograph or research paper to be heavily reliant on the codex form – surprisingly little changed with the advent of e-books and the PDF/browser-based journal article. These structures affect norms around argument, knowledge presentation, and the kinds of labor that are valued and those that are not. Clearly there is a great degree of heterogeneity within how individual authors, editors, and publishers across different disciplines utilize the form (which clearly affects an interdisciplinary project such as Ego Media). Nevertheless, the academic monograph, the gold standard of humanities-based scholarly communication, brings with it certain expressive affordances. Most notable is the expectation of what is communicable – a book-length argument that is built in a linear form. Although one can move nonsequentially around a codex, checking footnotes, entering via the index, and flipping back and forth, the author’s expectation is that the reader will anticipate beginning on the first page and continuing until the end of the book. This enables the author to develop an overall structure wherein individual chapters or sections build on those that come before, and to develop an argument that is linear, indeed even teleological, in its presentation of the author’s “critical interventions.” Such assumptions around structure dominate much academic writing today; indeed, they have the effect of painting the monograph as a strangely conservative form – the novel might be capacious enough to contain experiments in nonsequential reading such as those explored in Vladimir Nabokov’s Pale Fire or Mark Z. Danielewski’s Only Revolutions, but the experimental monograph is reconceived as creative nonfiction (Maggie Nelson’s The Argonauts or Maurice Blanchot’s The Writing of the Disaster spring to mind).

As a thought experiment: try reading the above paragraphs out of order. They are replete with internal signifiers of sequentiality, of a more or less explicit nature. The argument builds, contrasts, diverges, and returns in order to render itself persuasive. Reading nonsequentially is more than resistive reading here; in the case of the academic monograph to read it out of order is to reject the central premise of the form.

Academic monographs: Baked-in assumptions

The academic monograph also has baked within it certain assumptions about how and which data are presented. The emphasis on linearity might occasionally put pressure on narrative and description (depending on disciplinary norms), but it is the aphorism, the epiphany, the juxtaposition, and other such forms that are squeezed out. These modes of advancing thought, wherein the work of synthesis is displaced to the reader (unlike, for example, synthesis in description and argument, which build on each other within the prose)10 are not favored. So too, alternative forms of scholarship that focus on assembling, annotating, explicating, recovering, and alike are often crowded out in the modern scholarly monograph, and we risk devaluing these forms of knowledge transmission. The kinds of labor that are acknowledged and valued within academia are often tied to expectations around scholarly output: the work entailed in establishing – and maintaining – collaboration, collection, conversation, and collation is often gendered and discounted in favor of masculinist visions of the “charismatic” critic scripting their latest seismic intervention, subtle reading, or innovative theory, which will eventually be espoused via the monograph form.

In a digital setting such as this, codex-born structures and assumptions need not hold. Writers of digital fiction have explored the possibilities extensively – as the Electronic Literature Collections demonstrate.11 So too within digital humanities experiments in using mapping and timeline interfaces, crowdsourced contributions and editing, comment functions, and versioning have verified the potential of new forms of academic communication that break out of traditional codex-born structures. Yet despite such experimentation and despite our increasingly digital consumption of academic writing, the scholarly argument and critical intervention remain largely entrenched across the humanities. I have produced various examples, born of this research, myself.12,13,14 Here, I defy some of these norms in order to explore what we might learn.

This is more than a formal quibble. My interest in talking interfaces stems from an awareness that the conception of talk that they often model is a limited one – and the things that are left out or dampened are precisely those that can be so conducive to good communication and knowledge production. Talk can produce thought and feeling, it can enact speculation, clarification, and conjecture, it can create and constitute. It is an undervalued but central feature of academia – whether in the classroom, the conference hall, at the department meeting, or during a coffee break. Heterogeneous, unconstrained talk and the interactivity it is premised upon is vital to the good functioning of our discipline, just as it is to the good functioning of our society. In our rush to formalize talk in talking interfaces we too often delimit and prelimit these possibilities. In academia too we have privileged certain kinds of knowledge transmission above others, to our detriment. In writing of talking interfaces, I aim to render these actions visible (audible? sensible?) and thus potentially changeable.

What follows stages a series of questions both about talking interfaces (here the content) and about academic communication in a digital environment (here the form):

  • How might the affordances of the digital book be utilized to generate scholarship and what might this look like?
  • What kind of knowledge – and what aesthetics – can be communicated via circularity or more open-ended structures that we associate with talk?
  • What is the reader’s role in this environment and how might new processes of reading and response feed into knowledge production?
  • How might our understandings of existing talk interfaces and mediative norms be transformed in the process?

It is hoped these questions will guide the reader as you make use of the navigation options to choose your own route through this material. There is no right or wrong way to read; like a conversation, topics, themes, and questions are picked up and put down only to be taken up in other sections and discussed from different angles. Like a conversation, some things are left unsaid, anecdotes can do as much as argument, and juxtaposition, description, and narration work alongside critical exposition. You might find yourself reading at cross purposes or frustrated by a silence or repetition. As much as the error may be mine, be a kind reader and take such irritants as epistemologically suggestive. As much as the content, the form is the message here.

Pick a link to begin.

Literature, Talk, and Computing

Chatbots

Public Interfaces

Opacity and Not Talking

Endnotes

  1. I speak of a simple chatbot here; for more expansive discussion, see chatbots. ↩
  2. Exciting efforts to render digital scholarship more interactive continue apace, whether via the crowdsourcing efforts of Wikipedia, attempts to utilize reader feedback through versioning such as Kathleen Fitzpatrick’s Planned Obsolescence project on Media Commons Press at https://mcpress.media-commons.org/plannedobsolescence/, and alike. The AHRC-sponsored project The Academic Book of the Future at https://academicbookfuture.org/ has published a number of outputs on this topic. See also Tara McPherson, Feminist in a Software Lab: Difference + Design / Tara McPherson., MetaLABprojects (Cambridge, Mass.: Harvard University Press, 2018). ↩
  3. Angelika Bammer and Ruth-Ellen Boetcher Joeres, eds., The Future of Scholarly Writing: Critical Interventions (New York: Palgrave Macmillan, 2015), https://doi.org/10.1057/9781137505965. ↩
  4. Michael A. Elliott, “The Future of the Monograph in the Digital Era: A Report to the Andrew W. Mellon Foundation,” Journal of Electronic Publishing 18, no. 4 (2015), https://doi.org/https://doi.org/10.3998/3336451.0018.407. ↩
  5. There is an extensive body of work discussing ergodic and hypertext fiction, with early and influential texts including George P. Landow, Hypertext: The Convergence of Contemporary Critical Theory and Technology, Parallax (Baltimore, Md.: Johns Hopkins University Press, 1992). ↩
  6. J. David Bolter, Writing Space: The Computer, Hypertext, and the History of Writing (New York: Routledge, 1991). ↩
  7. Espen J. Aarseth, Cybertext: Perspectives on Ergodic Literature (Baltimore, Md.; London: Johns Hopkins University Press, 1997). ↩
  8. Nick Montfort, Twisty Little Passages: An Approach to Interactive Fiction (Cambridge, Mass.; London: MIT Press, 2003). ↩
  9. Peggy Phelan, Mourning Sex: Performing Public Memories (London: Routledge, 1997). 11–16. ↩
  10. Heather Love, “Close Reading and Thin Description,” Public Culture 25, no. 3 (71) (September 1, 2013): 401–34, https://doi.org/10.1215/08992363-2144688. ↩
  11. Electronic Literature Organization, Electronic Literature Collection, n.d., https://collection.eliterature.org/. ↩
  12. For those interventions, I encourage you to read Rebecca Roach, Literature and the Rise of the Interview (Oxford: Oxford University Press, 2018). ↩
  13. Rebecca Roach, “‘Three Words You Must Never Say:’ Hermione Lee on Interviewing,” Biography 41, no. 2 (2018): 270–86, https://doi.org/doi:10.1353/bio.2018.0023. ↩
  14. Rebecca Roach, “The Role and Function of Author Interviews in the Contemporary Anglophone Literary Field,” Book History 23 (2020): 335–64, https://doi.org/http://doi.org/10.1353/bh.2020.0009. ↩

Bibliography

  • Aarseth, Espen J. Cybertext: Perspectives on Ergodic Literature. Baltimore, Md.; London: Johns Hopkins University Press, 1997.
  • Bammer, Angelika, and Ruth-Ellen Boetcher Joeres, eds. The Future of Scholarly Writing: Critical Interventions. New York: Palgrave Macmillan, 2015. https://doi.org/10.1057/9781137505965.
  • Bolter, J. David. Writing Space: The Computer, Hypertext, and the History of Writing. New York: Routledge, 1991.
  • Electronic Literature Organization. Electronic Literature Collection, n.d. https://collection.eliterature.org/.
  • Elliott, Michael A. “The Future of the Monograph in the Digital Era: A Report to the Andrew W. Mellon Foundation.” Journal of Electronic Publishing 18, no. 4 (2015). https://doi.org/https://doi.org/10.3998/3336451.0018.407.
  • Love, Heather. “Close Reading and Thin Description.” Public Culture 25, no. 3 (71) (September 1, 2013): 401–34. https://doi.org/10.1215/08992363-2144688.
  • Landow, George P. Hypertext: The Convergence of Contemporary Critical Theory and Technology. Parallax. Baltimore, Md.: Johns Hopkins University Press, 1992.
  • McPherson, Tara. Feminist in a Software Lab: Difference + Design / Tara McPherson. MetaLABprojects. Cambridge, Mass.: Harvard University Press, 2018.
  • Montfort, Nick. Twisty Little Passages: An Approach to Interactive Fiction. Cambridge, Mass.; London: MIT Press, 2003.
  • Phelan, Peggy. Mourning Sex: Performing Public Memories. London: Routledge, 1997.
  • Roach, Rebecca. Literature and the Rise of the Interview. Oxford: Oxford University Press, 2018.
  • Roach, Rebecca. “The Role and Function of Author Interviews in the Contemporary Anglophone Literary Field.” Book History 23 (2020): 335–64. https://doi.org/http://doi.org/10.1353/bh.2020.0009.
  • Roach, Rebecca. “‘Three Words You Must Never Say:’ Hermione Lee on Interviewing.” Biography 41, no. 2 (2018): 270–86. https://doi.org/doi:10.1353/bio.2018.0023.